mne.io.read_info() of CTF .ds returns IsADirectoryError

[mjcourte]

Description of the problem

Hi all,

I have CTF 275 channel .ds, which reads normally with mne.io.read_raw_ctf().

When imported, the raw.info attribute 'appears' to be a complete mne.Info object, with about 19 keys.

However, mne.io.read_info() returns error: IsADirectoryError: [Errno 21] Is a directory 'path/to/my/data.ds',

When, later in script, I run mne.make_forward_solution(), which requires an mne.Info arg, I get the same IsADirectoryError while checking the inputs, if I pass raw.info (as opposed to trying a read_info() earlier). (Edit: this was a typo on my behalf)

While this is technically correct that the CTF .ds spec is a directory, is this expected behaviour? I don't see any specific docs about doing format conversion for CTF (eg .ds to .fif) to produce an info object.

Thanks, Matt

Steps to reproduce

import mne
from mne.datasets.brainstorm import bst_auditory

data_path = bst_auditory.data_path()
subject = 'bst_auditory'
subjects_dir = data_path / 'subjects'

raw_fname = data_path / "MEG" / subject / "S01_AEF_20131218_01.ds"

raw = mne.io.read_raw_ctf(raw_fname)
print("Info from read_raw_ctf raw.info:")
print(raw.info)

info = mne.io.read_info(raw_fname)

Link to data

No response

Expected results

mne.io.read_info of CTF data returns an mne.Info object, or, raw.info from mne.io.read_raw_ctf is interpreted as an mne.Info object

Actual results

In [8]: raw = mne.io.read_raw_ctf(raw_fname)
ds directory : /home/matt/mne_data/MNE-brainstorm-data/bst_auditory/MEG/bst_auditory/S01_AEF_20131218_01.ds
    res4 data read.
    hc data read.
    Separate EEG position data file read.
    Quaternion matching (desired vs. transformed):
       2.51   74.26    0.00 mm <->    2.51   74.26    0.00 mm (orig :  -56.69   50.20 -264.38 mm) diff =    0.000 mm
      -2.51  -74.26    0.00 mm <->   -2.51  -74.26    0.00 mm (orig :   50.89  -52.31 -265.88 mm) diff =    0.000 mm
     108.63    0.00    0.00 mm <->  108.63    0.00    0.00 mm (orig :   67.41   77.68 -239.53 mm) diff =    0.000 mm
    Coordinate transformations established.
    Reading digitizer points from ['/home/matt/mne_data/MNE-brainstorm-data/bst_auditory/MEG/bst_auditory/S01_AEF_20131218_01.ds/S01_20131218_01.pos']...
    Polhemus data for 3 HPI coils added
    Device coordinate locations for 3 HPI coils added
    5 extra points added to Polhemus data.
    Measurement info composed.
Finding samples for /home/matt/mne_data/MNE-brainstorm-data/bst_auditory/MEG/bst_auditory/S01_AEF_20131218_01.ds/S01_AEF_20131218_01.meg4: 
    System clock channel is available, checking which samples are valid.
    360 x 2400 = 864000 samples from 340 chs
Current compensation grade : 3
``

In [9]: print(raw.info)
<Info | 16 non-empty values
 bads: []
 ch_names: UDIO001, UPPT001, UTRG001, SCLK01-177, BG1-4408, BG2-4408, ...
 chs: 3 Stimulus, 32 misc, 26 Reference Magnetometers, 274 Magnetometers, 5 EEG
 comps: 5 items (list)
 ctf_head_t: CTF/4D/KIT head -> head transform
 custom_ref_applied: False
 dev_ctf_t: MEG device -> CTF/4D/KIT head transform
 dev_head_t: MEG device -> head transform
 dig: 263 items (3 Cardinal, 260 Extra)
 experimenter: EAB
 highpass: 0.0 Hz
 hpi_results: 1 item (list)
 lowpass: 1200.0 Hz
 meas_date: 2013-12-18 09:43:00 UTC
 meas_id: 4 items (dict)
 nchan: 340
 projs: []
 sfreq: 2400.0 Hz
 subject_info: 1 item (dict)
>

In [10]: info = mne.io.read_info(raw_fname)
---------------------------------------------------------------------------
IsADirectoryError                         Traceback (most recent call last)
Cell In[10], line 1
----> 1 info = mne.io.read_info(raw_fname)

File <decorator-gen-42>:12, in read_info(fname, verbose)

File ~/envs/mne/lib/python3.10/site-packages/mne/io/meas_info.py:1992, in read_info(fname, verbose)
   1978 @verbose
   1979 def read_info(fname, verbose=None):
   1980     """Read measurement info from a file.
   1981 
   1982     Parameters
   (...)
   1990     %(info_not_none)s
   1991     """
-> 1992     f, tree, _ = fiff_open(fname)
   1993     with f as fid:
   1994         info = read_meas_info(fid, tree)[0]

File <decorator-gen-10>:12, in fiff_open(fname, preload, verbose)

File ~/envs/mne/lib/python3.10/site-packages/mne/io/open.py:124, in fiff_open(fname, preload, verbose)
    100 @verbose
    101 def fiff_open(fname, preload=False, verbose=None):
    102     """Open a FIF file.
    103 
    104     Parameters
   (...)
    122         A list of tags.
    123     """
--> 124     fid = _fiff_get_fid(fname)
    125     try:
    126         return _fiff_open(fname, fid, preload)

File ~/envs/mne/lib/python3.10/site-packages/mne/io/open.py:52, in _fiff_get_fid(fname)
     50     else:
     51         logger.debug("Using normal I/O")
---> 52         fid = open(fname, "rb")  # Open in binary mode
     53 return fid

IsADirectoryError: [Errno 21] Is a directory: '/home/matt/mne_data/MNE-brainstorm-data/bst_auditory/MEG/bst_auditory/S01_AEF_20131218_01.ds'

Additional information

Windows 11/WSL2/Ubuntu 22.04 (such that FreeSurfer is accessible also within WSL)

Platform             Linux-5.15.90.1-microsoft-standard-WSL2-x86_64-with-glibc2.35
Python               3.10.12 (main, Jun 11 2023, 05:26:28) [GCC 11.4.0]
Executable           /home/matt/envs/mne/bin/python
CPU                  x86_64 (16 cores)
Memory               31.3 GB

Core
├☑ mne               1.5.1
├☑ numpy             1.26.0 (OpenBLAS 0.3.23.dev with 16 threads)
├☑ scipy             1.11.3
├☑ matplotlib        3.8.0 (backend=module://matplotlib_inline.backend_inline)
├☑ pooch             1.7.0
└☑ jinja2            3.1.2

Numerical (optional)
├☑ sklearn           1.3.1
├☑ nibabel           5.1.0
├☑ pandas            2.1.1
└☐ unavailable       numba, nilearn, dipy, openmeeg, cupy

Visualization (optional)
├☑ pyvistaqt         0.11.0
├☑ vtk               9.2.6
├☑ ipympl            0.9.3
├☑ mne-qt-browser    0.5.2
├☑ ipywidgets        8.1.1
├☑ trame_client      2.12.0
├☑ trame_server      2.12.0
├☑ trame_vtk         2.5.8
├☑ trame_vuetify     2.3.1
└☐ unavailable       pyvista, qtpy, pyqtgraph

Ecosystem (optional)
├☑ mne-connectivity  0.5.0
├☑ mne-icalabel      0.4
└☐ unavailable       mne-bids, mne-nirs, mne-features, mne-bids-pipeline

[larsoner]

read_info is really only meant to read FIF info files. You should be able to use

info = mne.io.read_raw_ctf(...).info

instead. If this makes sense would you be up for making a PR to improve the documentation of read_info to mention that it's designed for FIF info only but you can use code like the above to get info for other file formats?

[mjcourte]

I confirm that read_info works with sample_audvis_raw.fif.

I must have had a typo, as after going back to my script, the ctf raw.info appears to do the job.

I'll get to a PR for the read_info docs this afternoon.

[hoechenberger]

@larsoner Why don't we fail early if the filename extension is not .fif?

[larsoner]

Checking for .fif / .fif.gz is a good idea. I think check_fname could do that and emit a warning

[hoechenberger]

Why just a warning and not fail hard?

[larsoner]

Technically you can read and write FIF (and many other formats) just fine with no file extension. IIRC for many other functions like read raw fif etc we just emit a warning when you use non standard naming so we should continue that practice